www.nature.com/scientificreports
OPEN
received: 02 September 2014 accepted: 12 May 2015 Published: 05 June 2015
The mechanism for exclusion of Pinus massoniana during the succession in subtropical forest ecosystems: light competition or stoichiometric homoeostasis? Junhua Yan1, Kun Li1,3, Xingju Peng2, Zhongliang Huang1, Shizhong Liu1 & Qianmei Zhang1 Competition for light has traditionally been considered as the main mechanism for exclusion of Pinus massoniana during succession in subtropical forest ecosystems. However, both long-term inventories and a seedling cultivation experiment showed that growth of mature individuals and young seedlings of P. massoniana was not limited by available light, but was strongly influenced by stoichiometric homoeostasis. This is supported by the results of homoeostatic regulation coefficients for nitrogen (HN) and phosphorus (HP) estimated using the measured data from six transitional forests across subtropical China. Among three dominant tree species in subtropical forests, P. massoniana and Castanopsis chinensis had the lowest values of HP and HN, respectively. Therefore P. massoniana cannot survive in the advanced stage due to soil phosphorus limitation and C. chinensis cannot successfully grow in the pioneer stage due to soil nitrogen limitation. Our results support that stoichiometric homeostasis is the main reason for gradual exclusion of P. massoniana from the transitional forest and the eventual elimination from the advanced forest during the subtropical forest succession. Therefore greater attention should be paid to stoichiometric homeostasis as one of the key mechanisms for species exclusion during forest succession.
Competition and facilitation have significant effects on the structure and dynamics of ecosystems and therefore have been a focus in ecology1–3. Light competition has traditionally been considered as the main mechanism for tree species elimination from a forest ecosystem. Pinus massoniana is an evergreen conifer, native to many regions in the subtropical China (Fig. 1). That species are commonly used as the pioneering trees in forest recovery for degraded ecosystems in the subtropical China. Therefore, P. massoniana often dominates the pioneer stage and grows well in the acid red soils4. During the forest succession, P. massoniana declines in both plant biomass and abundance at the transitional stage, and disappears eventually at the advanced stage5. As the pioneer forests are invaded by native broadleaved species, such as Schima superb and Castanopsis chinensis that often grow faster than P. massoniana, and outcompete and eventually exclude P. massoniana from the forests, which is traditionally considered being the result of light competition6–8. However this has rarely been tested using field observations. From pioneer to transitional stage, large individuals of P. massoniana are initial dominant in the forest ecosystem, which intercept most incoming sunlight. Less sunlight is then available for the other invaded species which grow under the canopy of P. massoniana trees. The mechanism of light competition is 1
Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, China. 2Hennan University of Animal Husbandry and Economy, Zhengzhou 450046, China. 3University of Chinese Academy of Sciences, Beijing 100049, China. Correspondence and requests for materials should be addressed to J.Y. (email: [email protected]) Scientific Reports | 5:10994 | DOI: 10.1038/srep10994
1
www.nature.com/scientificreports/
Figure 1. Spatial distribution of the selected six transitional forests across the subtropical China. The data set is provided by Data Center of Resources and Environmental Sciences, Chinese Academy of Sciences (http://www.resdc.cn). Map showing the distribution was made using geographic information system software (ArcGIS version 9.3; ESRI 2012). At Dinghushan site, we have conducted a long-term forest inventory on the dynamics of tree species along the subtropical forest succession gradient since 1978 and an experiment of Pinus massoniana seedlings cultivation in the different succession stages since 2010.
unlikely to support exclusion of P. Massoniana under this condition. The other possible mechanism may be stoichiometric homeostasis, which characterizes the ability of an individual tree in maintaining a relatively steady nutrient balance in a variable environment9–11. Previous studies found that there were significant changes in soil biogeochemistry, particularly soil nitrogen (N) and phosphorus (P) availability from pioneer to advanced stage12,13. If the invaded tree species are more capable of maintaining stoichiometric homeostasis than P. massoniana under the soil conditions at transitional or advance stage, they will outcompete and eventually exclude P. massoniana during the forest succession. However these two mechanisms of species elimination operate fundamentally different at ecosystem scale. Competition for light depends on forest structure and shade tolerance of young seedlings, whereas stoichiometric homeostasis represents the ability of an individual organism to maintain its nutrient balance in highly variable soil environments, such as the changes in available soil N14–16 and soil P13,17–18 that have been observed during forest succession. To determine the primary mechanism for the exclusion of P. massoniana during a subtropical forest succession, we compiled long-term forest inventory data from the pioneer, transitional and advanced stages at Dinghushan Biosphere Reserve (DBR). These data were used to show the changes in biomass, individual number and tree height of P. massoniana during succession. We also collected data from an experiment that cultivated P. massoniana seedlings in a large forest gap within the pioneer, transitional or advanced forests at DBR. These data were used to assess whether regeneration of P. massoniana was mainly limited by sunlight or stoichiometric homoeostasis. Finally, we collected foliar N and P concentrations and available soil N and P from six transitional forests across subtropical China (Fig. 1). These data were used to calculate the homoeostatic regulation coefficients (H)11 of the dominated tree species in subtropical forests. The results from the three approaches will be used to explain why P. massoniana is gradually excluded from the transitional forest and eliminated from the advanced forest during forest succession.
Results
Changes in biomass fraction, individual number and tree height of P. massoniana. In the pio-
neer subtropical forest ecosystem, P. massoniana was the dominated tree species and occupied the entire forest canopy. Biomass of P. massoniana accounted for more than 80% of the ecosystem total (Table 1). The decrease in individual number of P. massoniana in the pioneer forest was a result of self-thinning, as there was no significant invasion by other tree species into the pioneer forest. During forest succession, the biomass fraction of P. massoniana decreased to 62% or 34% in the transitional forest and to zero in the advanced forest (Table 1). However the mean tree height of P. massoniana increased as tree grew bigger, and was always greater than other later successional tree species in the transitional forest Scientific Reports | 5:10994 | DOI: 10.1038/srep10994
2
www.nature.com/scientificreports/
Year
Transitional forest
Pioneer forest
Forest succession
Plot 1
Advanced forest
Plot 2
1978 1990 1995 2002 2006 2010 2002 2006 2010 2002 2006 2010 1992 1995 2002 2006 2010 n
465
465
465
395
350
258
242
175
d(cm) 10.8 14.0
16.2
19.2 20.9
22.4
26.0
26.7
27.6 28.2
29.5 34.1
h(m)
5.6
6.9
7.3
8.4
9.5
10.1
14.2
14.7
15.3 14.4
15.4 17.1
w(%)
100
100
100
98.4 92.5
82.3
74.9
65.7 61.9 51.1
45.8 34.1
n
417
425
275
142
Schima
d(cm)
9.3
13.4
14.1 17.9
superba
h(m)
6.5
9.2
w(%)
7.1
n
Pinus massoniana
850
179
92
37
35
33
18.9 21.8
30.3
32.1
34.6
36.3 37.6
10.0 11.6
11.7 14.6
18.0
18.4
18.7
19.4 21.2
10.4
10.4 12.5
13.7 16.4
18.3
21.7
22.2
22.3 24.8
300
283
250
192
14
14
12
Castanopsis d(cm)
9.6
10.2
11.1 19.1
20.4 23.5
79.2
80.3
82.2
84.6 88.1
h(m)
6.3
7.1
7.4
11.6
12.0 13.9
22.9
22.9
23.5
25.7 28.9
w(%)
10.5
15.6
18.7 26.2
28.2 35.4
53.4
52.8
56.8
chinensis
158
167
192
142
183
31
11
57
26
7
49.9
n
136
158
233
616
758
700
Others
d(cm)
5.1
10.9
15.0
7.2
7.4
7.4
6.9
7.0
7.1
10.2
10.5
9.1
7.9
7.8
(h>5m)
h(m)
6.1
7.1
7.3
7.3
7.5
7.3
7.2
7.3
7.9
8.7
8.7
7.7
7.9
7.8
w(%)
1.6
7.5
17.7
7.5
8.3
9.0
10.2
12.3 14.1
28.3
25.5
21
20.7 25.3
1650 1492 1267 1266 1233 1117 1484 1681
Table 1. Description of dominant tree species in a permanent plot with 1ha at Dinghushan Biosphere Reserve. The pioneer or advanced forest has one plot. The transitional forest has two plots, one is located in buffer area (Plot 1) and the other is located in core area (Plot 2) of the reserve. Plot 2 has not been disturbed longer than plot 1. n is individual number; d and h are mean diameter and height of tree species, respectively; w is the fraction of tree species biomass in total tree biomass.
(Table 1). Therefore above-ground light competition is unlikely to explain why P. massoniana decreased its biomass and became extinct in the advanced forest.
Is the regeneration of P. massoniana light limited? Although above-ground light competition among mature trees cannot explain the decreases in biomass and individual number of P. massoniana during forest succession, it is possible that P. massoniana species did not successfully regenerate because of light limitation under the canopy, therefore were gradually excluded from the transitional forest as old P. massoniana trees died. To test this hypothesis, we cultivated 100 seedlings of P. massoniana in a large forest gap receiving full sunlight in each of the three successional forests at DBR over four years (Fig. 2). In the first year, seedlings suffered 7%, 13% and 13% mortality in the pioneer, transitional and advanced forest gaps, respectively. Seedlings mortality rate increased greatly in the second year, with mortality rate of 15%, 44% and 63% in the pioneer, transitional and advanced forest gaps, respectively. In the transitional and advanced forests, all 100 seedlings died by the fourth year, whereas 67% seedlings of P. massoniana survived in the pioneer forest at the end of the experiment. Therefore regeneration of P. massoniana in the transitional or advanced forest was not limited by light. As shown in Table 2, measurements of foliar N and P concentrations show that foliar N concentration was highest and foliar P concentration was lowest in the seedlings of P. massoniana in the forest gap of the advanced forest. The differences in foliar P concentration between the advanced or transitional forest and pioneer forest were statistically significant (Table 2). As a result, foliar N:P ratio was lowest in the pioneer forest and highest in the advanced forest, and all differences in the seedlings among three forest gaps are statistically significant (Table 2). Comparing H of three dominant tree species across an environmental gradient. The measured foliar N and P concentrations of P. massoniana mature trees and available soil N and P concentrations at Scientific Reports | 5:10994 | DOI: 10.1038/srep10994
3
www.nature.com/scientificreports/
Figure 2. Number of survial Pinus massoniana seedlings that cultivated in pioneer, transitional and advanced forest stands along the subtropical forest succession gradient at Dinghushan Biosphere Reserve.
Forest
N (mg g−1)
P (mg g−1)
N:P ratio
Pioneer
20.79 ± 1.68
a
1.43 ± 0.21a
14.08 ± 1.86a
Transitional
21.54 ± 1.38a
0.99 ± 0.08b
21.76 ± 1.67b
Advanced
22.10 ± 1.30
0.87 ± 0.08
25.40 ± 1.93c
a
b
Table 2. Foliar N and P stoichiometry of P. massoniana seedlings that cultivated in pioneer, transitional and advanced forest stands along the subtropical forest succession gradient at Dinghushan Biophere Reserve. The mean ± standard deviation followed by different lowercase letters has significant differences (p
OPEN
received: 02 September 2014 accepted: 12 May 2015 Published: 05 June 2015
The mechanism for exclusion of Pinus massoniana during the succession in subtropical forest ecosystems: light competition or stoichiometric homoeostasis? Junhua Yan1, Kun Li1,3, Xingju Peng2, Zhongliang Huang1, Shizhong Liu1 & Qianmei Zhang1 Competition for light has traditionally been considered as the main mechanism for exclusion of Pinus massoniana during succession in subtropical forest ecosystems. However, both long-term inventories and a seedling cultivation experiment showed that growth of mature individuals and young seedlings of P. massoniana was not limited by available light, but was strongly influenced by stoichiometric homoeostasis. This is supported by the results of homoeostatic regulation coefficients for nitrogen (HN) and phosphorus (HP) estimated using the measured data from six transitional forests across subtropical China. Among three dominant tree species in subtropical forests, P. massoniana and Castanopsis chinensis had the lowest values of HP and HN, respectively. Therefore P. massoniana cannot survive in the advanced stage due to soil phosphorus limitation and C. chinensis cannot successfully grow in the pioneer stage due to soil nitrogen limitation. Our results support that stoichiometric homeostasis is the main reason for gradual exclusion of P. massoniana from the transitional forest and the eventual elimination from the advanced forest during the subtropical forest succession. Therefore greater attention should be paid to stoichiometric homeostasis as one of the key mechanisms for species exclusion during forest succession.
Competition and facilitation have significant effects on the structure and dynamics of ecosystems and therefore have been a focus in ecology1–3. Light competition has traditionally been considered as the main mechanism for tree species elimination from a forest ecosystem. Pinus massoniana is an evergreen conifer, native to many regions in the subtropical China (Fig. 1). That species are commonly used as the pioneering trees in forest recovery for degraded ecosystems in the subtropical China. Therefore, P. massoniana often dominates the pioneer stage and grows well in the acid red soils4. During the forest succession, P. massoniana declines in both plant biomass and abundance at the transitional stage, and disappears eventually at the advanced stage5. As the pioneer forests are invaded by native broadleaved species, such as Schima superb and Castanopsis chinensis that often grow faster than P. massoniana, and outcompete and eventually exclude P. massoniana from the forests, which is traditionally considered being the result of light competition6–8. However this has rarely been tested using field observations. From pioneer to transitional stage, large individuals of P. massoniana are initial dominant in the forest ecosystem, which intercept most incoming sunlight. Less sunlight is then available for the other invaded species which grow under the canopy of P. massoniana trees. The mechanism of light competition is 1
Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, China. 2Hennan University of Animal Husbandry and Economy, Zhengzhou 450046, China. 3University of Chinese Academy of Sciences, Beijing 100049, China. Correspondence and requests for materials should be addressed to J.Y. (email: [email protected]) Scientific Reports | 5:10994 | DOI: 10.1038/srep10994
1
www.nature.com/scientificreports/
Figure 1. Spatial distribution of the selected six transitional forests across the subtropical China. The data set is provided by Data Center of Resources and Environmental Sciences, Chinese Academy of Sciences (http://www.resdc.cn). Map showing the distribution was made using geographic information system software (ArcGIS version 9.3; ESRI 2012). At Dinghushan site, we have conducted a long-term forest inventory on the dynamics of tree species along the subtropical forest succession gradient since 1978 and an experiment of Pinus massoniana seedlings cultivation in the different succession stages since 2010.
unlikely to support exclusion of P. Massoniana under this condition. The other possible mechanism may be stoichiometric homeostasis, which characterizes the ability of an individual tree in maintaining a relatively steady nutrient balance in a variable environment9–11. Previous studies found that there were significant changes in soil biogeochemistry, particularly soil nitrogen (N) and phosphorus (P) availability from pioneer to advanced stage12,13. If the invaded tree species are more capable of maintaining stoichiometric homeostasis than P. massoniana under the soil conditions at transitional or advance stage, they will outcompete and eventually exclude P. massoniana during the forest succession. However these two mechanisms of species elimination operate fundamentally different at ecosystem scale. Competition for light depends on forest structure and shade tolerance of young seedlings, whereas stoichiometric homeostasis represents the ability of an individual organism to maintain its nutrient balance in highly variable soil environments, such as the changes in available soil N14–16 and soil P13,17–18 that have been observed during forest succession. To determine the primary mechanism for the exclusion of P. massoniana during a subtropical forest succession, we compiled long-term forest inventory data from the pioneer, transitional and advanced stages at Dinghushan Biosphere Reserve (DBR). These data were used to show the changes in biomass, individual number and tree height of P. massoniana during succession. We also collected data from an experiment that cultivated P. massoniana seedlings in a large forest gap within the pioneer, transitional or advanced forests at DBR. These data were used to assess whether regeneration of P. massoniana was mainly limited by sunlight or stoichiometric homoeostasis. Finally, we collected foliar N and P concentrations and available soil N and P from six transitional forests across subtropical China (Fig. 1). These data were used to calculate the homoeostatic regulation coefficients (H)11 of the dominated tree species in subtropical forests. The results from the three approaches will be used to explain why P. massoniana is gradually excluded from the transitional forest and eliminated from the advanced forest during forest succession.
Results
Changes in biomass fraction, individual number and tree height of P. massoniana. In the pio-
neer subtropical forest ecosystem, P. massoniana was the dominated tree species and occupied the entire forest canopy. Biomass of P. massoniana accounted for more than 80% of the ecosystem total (Table 1). The decrease in individual number of P. massoniana in the pioneer forest was a result of self-thinning, as there was no significant invasion by other tree species into the pioneer forest. During forest succession, the biomass fraction of P. massoniana decreased to 62% or 34% in the transitional forest and to zero in the advanced forest (Table 1). However the mean tree height of P. massoniana increased as tree grew bigger, and was always greater than other later successional tree species in the transitional forest Scientific Reports | 5:10994 | DOI: 10.1038/srep10994
2
www.nature.com/scientificreports/
Year
Transitional forest
Pioneer forest
Forest succession
Plot 1
Advanced forest
Plot 2
1978 1990 1995 2002 2006 2010 2002 2006 2010 2002 2006 2010 1992 1995 2002 2006 2010 n
465
465
465
395
350
258
242
175
d(cm) 10.8 14.0
16.2
19.2 20.9
22.4
26.0
26.7
27.6 28.2
29.5 34.1
h(m)
5.6
6.9
7.3
8.4
9.5
10.1
14.2
14.7
15.3 14.4
15.4 17.1
w(%)
100
100
100
98.4 92.5
82.3
74.9
65.7 61.9 51.1
45.8 34.1
n
417
425
275
142
Schima
d(cm)
9.3
13.4
14.1 17.9
superba
h(m)
6.5
9.2
w(%)
7.1
n
Pinus massoniana
850
179
92
37
35
33
18.9 21.8
30.3
32.1
34.6
36.3 37.6
10.0 11.6
11.7 14.6
18.0
18.4
18.7
19.4 21.2
10.4
10.4 12.5
13.7 16.4
18.3
21.7
22.2
22.3 24.8
300
283
250
192
14
14
12
Castanopsis d(cm)
9.6
10.2
11.1 19.1
20.4 23.5
79.2
80.3
82.2
84.6 88.1
h(m)
6.3
7.1
7.4
11.6
12.0 13.9
22.9
22.9
23.5
25.7 28.9
w(%)
10.5
15.6
18.7 26.2
28.2 35.4
53.4
52.8
56.8
chinensis
158
167
192
142
183
31
11
57
26
7
49.9
n
136
158
233
616
758
700
Others
d(cm)
5.1
10.9
15.0
7.2
7.4
7.4
6.9
7.0
7.1
10.2
10.5
9.1
7.9
7.8
(h>5m)
h(m)
6.1
7.1
7.3
7.3
7.5
7.3
7.2
7.3
7.9
8.7
8.7
7.7
7.9
7.8
w(%)
1.6
7.5
17.7
7.5
8.3
9.0
10.2
12.3 14.1
28.3
25.5
21
20.7 25.3
1650 1492 1267 1266 1233 1117 1484 1681
Table 1. Description of dominant tree species in a permanent plot with 1ha at Dinghushan Biosphere Reserve. The pioneer or advanced forest has one plot. The transitional forest has two plots, one is located in buffer area (Plot 1) and the other is located in core area (Plot 2) of the reserve. Plot 2 has not been disturbed longer than plot 1. n is individual number; d and h are mean diameter and height of tree species, respectively; w is the fraction of tree species biomass in total tree biomass.
(Table 1). Therefore above-ground light competition is unlikely to explain why P. massoniana decreased its biomass and became extinct in the advanced forest.
Is the regeneration of P. massoniana light limited? Although above-ground light competition among mature trees cannot explain the decreases in biomass and individual number of P. massoniana during forest succession, it is possible that P. massoniana species did not successfully regenerate because of light limitation under the canopy, therefore were gradually excluded from the transitional forest as old P. massoniana trees died. To test this hypothesis, we cultivated 100 seedlings of P. massoniana in a large forest gap receiving full sunlight in each of the three successional forests at DBR over four years (Fig. 2). In the first year, seedlings suffered 7%, 13% and 13% mortality in the pioneer, transitional and advanced forest gaps, respectively. Seedlings mortality rate increased greatly in the second year, with mortality rate of 15%, 44% and 63% in the pioneer, transitional and advanced forest gaps, respectively. In the transitional and advanced forests, all 100 seedlings died by the fourth year, whereas 67% seedlings of P. massoniana survived in the pioneer forest at the end of the experiment. Therefore regeneration of P. massoniana in the transitional or advanced forest was not limited by light. As shown in Table 2, measurements of foliar N and P concentrations show that foliar N concentration was highest and foliar P concentration was lowest in the seedlings of P. massoniana in the forest gap of the advanced forest. The differences in foliar P concentration between the advanced or transitional forest and pioneer forest were statistically significant (Table 2). As a result, foliar N:P ratio was lowest in the pioneer forest and highest in the advanced forest, and all differences in the seedlings among three forest gaps are statistically significant (Table 2). Comparing H of three dominant tree species across an environmental gradient. The measured foliar N and P concentrations of P. massoniana mature trees and available soil N and P concentrations at Scientific Reports | 5:10994 | DOI: 10.1038/srep10994
3
www.nature.com/scientificreports/
Figure 2. Number of survial Pinus massoniana seedlings that cultivated in pioneer, transitional and advanced forest stands along the subtropical forest succession gradient at Dinghushan Biosphere Reserve.
Forest
N (mg g−1)
P (mg g−1)
N:P ratio
Pioneer
20.79 ± 1.68
a
1.43 ± 0.21a
14.08 ± 1.86a
Transitional
21.54 ± 1.38a
0.99 ± 0.08b
21.76 ± 1.67b
Advanced
22.10 ± 1.30
0.87 ± 0.08
25.40 ± 1.93c
a
b
Table 2. Foliar N and P stoichiometry of P. massoniana seedlings that cultivated in pioneer, transitional and advanced forest stands along the subtropical forest succession gradient at Dinghushan Biophere Reserve. The mean ± standard deviation followed by different lowercase letters has significant differences (p
